sciencebriefs
13:00in productionCh. 1 · A pathway that stays sensitised/ 13:00 · ceiling 15 min
Neuroscience

Long-term potentiation

In 1973 Timothy Bliss and Terje Lømo showed that briefly stimulating a pathway in a rabbit's brain left its synapses firing harder for hours afterward — physical evidence for Donald Hebb's 1949 idea that repeated use is what strengthens a connection between neurons.

Working in Per Andersen's laboratory in Oslo, Terje Lømo first noticed in 1966 that brief, intense electrical stimulation of a pathway feeding into the rabbit hippocampus left the synapses there responding more strongly long afterwards. Timothy Bliss joined the work in 1968, and their 1973 paper gave the effect its first formal description; it was later named long-term potentiation. The finding supplied a physical mechanism for Donald Hebb's older proposal that neurons which fire together end up wired together, and decades of subsequent work have tied it to spatial learning, addiction and Alzheimer's disease, while leaving open exactly how the cellular changes it produces add up to a stored memory.

Chapters & takeaways6
  1. 0:08
    A pathway that stays sensitised

    Lømo found that a brief burst of stimulation to a rabbit's hippocampal pathway left it responding more strongly for hours.

  2. 2:10
    Naming it, formally

    Bliss and Lømo's 1973 paper described the effect in the anaesthetised rabbit; the label long-term potentiation followed two years later.

  3. 4:20
    A mechanism for Hebb's rule

    Donald Hebb had proposed in 1949 that repeated joint activity strengthens a synapse; this gave that proposal something measurable to point to.

  4. 6:30
    From synapse to receptor

    Later work traced the effect to calcium entering through NMDA receptors and more AMPA receptors moving into the synapse.

  5. 8:40
    Blocking it, and watching learning fail

    Richard Morris showed in 1986 that drugging the NMDA receptor stopped both potentiation in hippocampal slices and rats' ability to learn a water maze.

  6. 10:50
    Necessary, not sufficient

    Even the researchers who mapped its molecular detail treat potentiation as one ingredient in memory, not proof of the whole recipe.

Worth your time?

Yes. Study the whole thing.

4/ 5
What works
  • the connection from a 1949 theoretical proposal to a 1973 measurable effect is genuinely satisfying to trace
  • the behavioural experiments, especially blocking the receptor and watching learning fail alongside it, make the link concrete
  • the source is candid that the effect is necessary rather than sufficient for memory, which keeps the claim honest
What does not
  • a molecule once thought essential to maintaining the effect, PKMζ, turned out not to be required after all, and the field has not fully replaced that explanation
  • how the cellular change actually becomes a stored, retrievable memory across a network of neurons is still not resolved
Study it if
  • anyone who wants the actual cellular evidence behind cells that fire together wire together
  • readers following why Alzheimer's and addiction research keep returning to this one hippocampal effect
  • anyone comfortable with a discovery that took decades to half-explain, rather than one tidy afternoon
Skip it if
  • readers wanting a settled account of how memories are stored, rather than one ingredient in it
  • anyone after a single clean experiment rather than a fifty-year accumulation of results
The written brief4 min read

A pathway that stays sensitised

The claim is that a synapse can be left more sensitive after brief, intense use, and that this change lasts far longer than the stimulation itself. Terje Lømo first observed it in 1966, running electrical stimulation into the perforant pathway of an anaesthetised rabbit and recording the response in the dentate gyrus, a part of the hippocampus. A short burst of high-frequency stimulation left the recorded response larger than before, and that enlargement persisted well beyond the stimulation. Timothy Bliss, who joined the Oslo laboratory in 1968, worked with Lømo to characterise the effect properly, and their 1973 paper gave the first full account of what would later be named long-term potentiation, now treated as the leading cellular candidate for how the brain stores what it learns.

Naming it, formally

The original work used anaesthetised rabbits and direct electrical stimulation of a specific hippocampal circuit, the perforant pathway feeding into the dentate gyrus, with electrodes recording the size of the postsynaptic response before and after a brief tetanic burst. That electrophysiological method, stimulate and record, remains the basic approach for studying the effect, extended over the following decades to hippocampal slices kept alive outside the animal and to genetically modified mice. Molecular work built on top of that method traced an early, short-lived phase driven by calcium entering postsynaptic cells through NMDA receptors and existing receptors being made more responsive, and a later phase, requiring new protein production, that involves physical growth in the connecting structures between neurons.

A mechanism for Hebb’s rule

What has held up is the basic phenomenon and its dependence, in many though not all pathways, on the NMDA receptor. Blocking that receptor with the drug APV, as Richard Morris did in 1986, stops potentiation from being produced in hippocampal tissue and, in the same animals, stops them learning to navigate a water maze, tying the cellular effect to a measurable behaviour rather than leaving it as a laboratory curiosity. Later experiments extended the case: mice engineered to carry an extra receptor subunit showed both enhanced potentiation and better spatial learning, and rats trained on an avoidance task showed the same receptor changes in their synapses as rats given the artificial stimulation directly, with previously trained synapses resistant to further potentiation in a way that mirrors the laboratory effect closely.

From synapse to receptor

What has not held up as cleanly is any single, complete molecular account of how the change is maintained over time. A protein called PKMζ was for years treated as essential to keeping the late phase of potentiation going, but mice bred without it still showed normal potentiation, undercutting a mechanism that had looked settled. Whether the presynaptic side of the connection, not just the postsynaptic side, contributes to the lasting change is still disputed, and researchers disagree about what signal, if any, would need to travel backward across the synapse to produce it. Where in the cell the new proteins needed for the lasting phase are actually made, near the synapse itself or back in the cell body, is also unresolved, meaning several of the field’s mechanistic claims remain provisional rather than settled.

Blocking it, and watching learning fail

The effect matters beyond the hippocampal slice because it gives researchers a physical target for conditions defined, until now, almost entirely by behaviour. Amyloid-beta, the protein fragment implicated in Alzheimer’s disease, has been shown to interfere with potentiation in the hippocampus, offering one candidate explanation for the memory decline that marks the disease early on. Addiction researchers have started treating compulsive drug use as a powerful, unwanted form of the same learning process, pointing to potentiation-like changes in the reward circuitry of the brain rather than in the hippocampus. Neither line of work amounts to a treatment yet, but both depend on the idea, established by Bliss and Lømo’s original result, that learning and memory have a traceable cellular signature rather than being purely a matter of psychology.

Necessary, not sufficient

Yes, though it rewards patience rather than delivering a single dramatic payoff. What makes it worth the time is watching a simple electrophysiological observation in an anaesthetised rabbit grow, over five decades, into a framework that touches memory, addiction and neurodegenerative disease, while never quite closing the gap between a stronger synapse and an actual stored memory. Readers who want certainty should know going in that some of the field’s earlier landmark explanations, the role of one particular maintenance protein among them, have since been walked back rather than confirmed. That openness is part of what makes the subject credible: this is a case where the caveats are as informative as the headline result, and worth reading past.

Same field · Neuroscience4 of 45
Up next in Science

Mirror neuron

· 13:00

In the early 1990s a Parma laboratory found monkey brain cells that fired both when the animal grasped food and when it merely watched someone else grasp it — a finding since stretched into theories of empathy, language and autism that the evidence itself struggles to support.

13:00